########################################################################################
Codebook for "Institutional Constraints on the Executive, Investment, and Elections" 
########################################################################################

List of replication code files: 
"main_analyses_and_supplemental_tables.do"; "footnote15_analysis.do";

List of datasets: 
"main_analyses_and_supplemental_tables.dta"; "POLCON_2017.dta"


#################################################################
Replication code file: "main_analyses_and_supplemental_tables.do"
#################################################################

// The code in this file replicates all analyses included in the article and the tables in the online supplemental materials, with the exception of analyses discussed in footnote 15. Those can be replicated using the replication code file "footnote15_analysis.do" (please see below).


#################################################
Replication code file: "footnote15_analysis.do"
#################################################

The code replicates the analyses discussed in footnote 15 of the article. In particular, it produces comparisons of within-country change in PolConV, as well as changes in the number of effective/independent institutions between the sample of OECD countries and the sample of countries used in the main analysis.


####################################################
Dataset: "main_analyses_and_supplemental_tables.dta"
####################################################

The dataset can be used to replicate all but one of the analyses included in the article and online supplemental materials. (The remaining analysis also uses "POLCON_2017.dta") The variables and variable descriptions for "main_analyses_and_supplemental_tables.dta" are included below. 


Variable: "countryname"

Identifies the name of the country for each country-year observation.


Variable: "countryid" 

ID variable to identify the country of the country-year observation ; panel variable for STATA analysis. 


Variable: "regionid"

ID variable to identify the region of the country-year observation. 

1: Africa
2: Asia
3: Europe
4: Central and South America


Variable: "year"

Variable to identify the year the country-year observation ; time variable for STATA analysis.


Variable: "obs_id"

ID variable giving a unique number to each observation.


Variable: "elt"

Executive Election variable ; based on e.g., Alesina, Cohen, and Roubini (1993) 

1: an election directly or indirectly determining the head executive occurred in the second half of year t or the first half of year t+1
0: otherwise


Variable: "e1" 

Executive Election - First Half-Year ; based on Brender and Drazen (2005) 

1: an election directly or indirectly determining the head executive occurred in the first half of the calendar year
0: otherwise


Variable: "e2" 

Executive Election - Second Half-Year ; based on Brender and Drazen (2005) 

1: an election directly or indirectly determining the head executive occurred in the second half of the calendar year
0: otherwise


Variable: "calelecyr"

Executive Election - Calendar Year

0: no election directly or indirectly determining the head executive during the calendar year
1: an election directly or indirectly determining the head executive during the calendar year


Variable: "termover"

Term Expires

1: term scheduled to expire in that year
0: otherwise


Variable: "polconv"

Institutional Constraints ; variable "PolConV" from Henisz (2017)


Variable: "lnpolconv" 

ln Institutional Constraints ; variable "polconv" logged according to the following formula: ln((polconv*100)+1)


Variable: "eltlnpolconv"

Executive Election x ln Institutional Constraints ; interaction between variables "elt" and "lnpolconv" (elt * lnpolconv)


Variable: "eltpolconv"

Executive Election x Institutional Constraints ; interaction between variables "elt" and "polconv" (elt * polconv)


Variable: "e1lnpolconv" 

Executive Election - First Half-Year x ln Institutional Constraints ; interaction between variables "e1" and "lnpolconv" (e1 * lnpolconv)


Variable: "e2lnpolconv"

Executive Election - Second Half-Year x ln Institutional Constraints ; interaction between variables "e2" and "lnpolconv" (e2 * lnpolconv)


Variable: "e1polconv" 

Executive Election - First Half-Year x Institutional Constraints ; interaction between variables "e1" and "polconv" (e1 * polconv)


Variable: "e2polconv"

Executive Election - Second Half-Year x Institutional Constraints ; interaction between variables "e2" and "polconv" (e2 * polconv)


Variable: "calelecyrpolconv"

Executive Election - Calendar Year x Institutional Constraints ; interaction between variables "calelecyr" and "polconv" (calelecyr * polconv)


Variable: "termoverpolconv"

Term Expires x Institutional Constraints ; interaction between variables "termover" and "polconv" (termover * polconv)


Variable: "termoverlnpolconv"

Term Expires x ln Institutional Constraints ; interaction between variables "termover" and "lnt1" (termover * lnpolconv)


Variable: "calelecyrlnpolconv"

Executive Election - Calendar Year x ln Institutional Constraints ; Interaction between variables "calelecyr" and "lnpolconv" (calelecyr * lnpolconv)


Variable: "xconst" 

Executive Constraints ; variable "XCONST" from Marshall, Gurr, and Jaggers (2019)


Variable: "lnxconst"

ln XCONST ;  variable "xconst" logged according to the following formula: ln(xconst)


Variable: "eltlnxconst" 

Executive Election x ln XCONST ; interaction between variables "elt" and "lnxconst" (elt * lnxconst)


Variable: "eltxconst"

Executive Election x XCONST ; interaction between variables "elt" and "xconst" (elt * xconst)


Variable: "v2x_liberal"
 
Liberal component index ; variable "v2x_liberal" from V-Dem 10 dataset (Coppedge et al. 2020)


Variable: "lnv2xlib" 

ln Liberal component index; variable "v2x_liberal" logged according to the following formula: ln(100*v2x_liberal)


Variable: "eltlnv2xlib"

Executive Election x ln Liberal component index ; interaction between variables "elt" and "lnv2xlib" (elt * lnv2xlib)


Variable: "eltv2xlib"

Executive Election x Liberal component index ; interaction between variables "elt" and "v2x_liberal" (elt * v2x_liberal)


Variable: "v2xel_frefair"

Free and fair elections ; variable "v2xel_frefair" from V-Dem 10 dataset (Coppedge et al. 2020)


Variable: "lnv2xelfrefair"

ln Free & fair elections ; variable "v2xel_frefair" logged according to the following formula: ln((100*v2xel_frefair)+1)


Variable: "eltlnv2xelfrefair"

Executive Election x ln Free and fair elections ; interaction between variables "elt" and "lnv2xelfrefair" (elt * lnv2xelfrefair)


Variable: "eltv2xelfrefair"

Executive Election x Free and fair elections ; interaction between variables "elt" and "v2xel_frefair" (elt * v2xel_frefair)


Variable: "c_mldi"

ML democracy ; continuous Machine Learning Democracy Indicator variables "c_mldi" from Gründler and Krieger (2021)


Variable: "lncmldi"

ln ML democracy ; variable "c_mldi" logged according to the following formula: ln((100*c_mldi)+1)


Variable: "eltlncmldi"

Executive Election x ln ML democracy ; interaction between variables "elt" and "lncmldi" (elt * lncmldi)


Variable: "eltcmldi"

Executive Election x ML democracy ; interaction between variables "elt" and "c_mldi" (elt * c_mldi)


Variable: "colony"

Colonized by European nations between 1500-1900. ID variable indicating whether the country was formerly a British, French, German, Spanish, Italian, Belgian, Dutch, or Portuguese colony ; based on Acemoglu, Johnson, and Robinson (2002) and Acemoglu and Johnson (2005)

Variable: "englishlaw" 

Legal origin ; variable indicating whether the country's legal system is based on the English Common Law system / coded by whether the country was formerly a British colony ; see Acemoglu, Johnson, and Robinson (2002) and Acemoglu and Johnson (2005)

Variable: "eltenglish" 

Executive Election x Legal Origin ; interaction between variables "elt" and "englishlaw" (elt * englishlaw)
 

Variable: "lpd1500s" 

ln Population density in 1500 ; natural log of population density in 1500 (see Acemoglu, Johnson, and Robinson 2002; Acemoglu and Johnson 2005)


Variable: "eltlpd1500s"

Executive Election x ln Population density in 1500 ; interaction between variables "elt" and "lpd1500s" (elt x lpd1500s)


Variable: "execrlc"

Ideology of government ; variable "EXECRLC" from DPI 2017 (Cruz, Keefer, and Scartascini 2018)

0: Non-ideological / not coded as ideological / NA
1: Right 
2: Center
3: Left


Variable: "regimedum1"

Ideology of government: Non-ideological / not coded as ideological / NA ; based on variable "EXECRLC"==0 (Cruz, Keefer, and Scartascini 2018)

1: Non-ideological / not coded as ideological / NA 
0: Government ideology coded as Center, Left, or Right


Variable: "regimedum2"
 
Right government ; based on variable "EXECRLC"==1 (Cruz, Keefer, and Scartascini 2018)

1: Government ideology coded as Right
0: Otherwise


Variable: "regimedum3" 

Center government; based on variable "EXECRLC"==2 (Cruz, Keefer, and Scartascini 2018)

1: Government ideology coded as Center
0: Otherwise


Variable: "regimedum4" 

Left government ; based on variable "EXECRLC"==3 (Cruz, Keefer, and Scartascini 2018)

1: Government ideology coded as Left
0: Otherwise


Variable: "rpt_for_elt"

Rational partisan theory (for variable "elt") ; variable based on, e.g., Alesina, Roubini, and Cohen (1997). 

-1: Government shifts from left to right in the year after the election, where the election is measured with the variable "elt"
1: Government shifts from right to left in the year after the election, where the election is measured with the variable "elt" 
0: all other cases


Variable: "rpt2"

Rational partisan theory (for other election indicators); variable based on, e.g., Alesina, Roubini, and Cohen (1997). 

-1: Government shifts from left to right in the year after the election, where the election is measured with the election variable "calelecyr"
1: Government shifts from right to left in the year after the election, where the election is measured with the election variable "calelecyr" 
0: all other cases


Variable: "privgfcf"

Gross fixed capital formation, private sector (current LCU) ; series name "NE.GDI.FPRV.CN" (Version Code: 201811) from World Development Indicators (see https://databank.worldbank.org/source/world-development-indicators)


Variable: "chrprivgfcf100_w005e, winsorized main variable"
 
Private fixed investment growth (%). The variable is first calculated as year-over-year change (x-x[_n-1])/x[_n-1] in real-valued private gross fixed capital formation ("privgfcf" adjusted by "cpi") in a given country, multiplied by 100. Then, winsorization is two-sided based on the fraction of 0.005 in each tail, based on the observations when main_sample==1. 


Variable: "chrprivgfcf100"

Private fixed investment growth (%), unwinsorized. 

The variable is calculated as year-over-year change (x-x[_n-1])/x[_n-1] in real-valued private gross fixed capital formation in a given country, multiplied by 100. 


Variable: "lngrowth"

ln Private fixed investment growth (%); logged Private fixed investment growth (%)

The variable chrprivgfcf100 is transformed according to following formula: ln(chrprivgfcf100+76); based on the minimum value of "chrprivgfcf100" when main_sample==1. 


Variable: "lngrowthIHS"

Private fixed investment growth (%), IHS log-transformation. 

The variable chrprivgfcf100 is transformed according to following formula:  ln((sqrt(chrprivgfcf100^2)+1)+chrprivgfcf100). 


Variable: "realpcgdpUS"

GDP per capita (constant 2010 US$) ; series name "NY.GDP.PCAP.KD" (Version Code: 201811) from World Development Indicators (see https://databank.worldbank.org/source/world-development-indicators)


Variable: "chlgrealpcgdpUS_w005e"

lagged per capita GDP growth (in $US), winsorized main variable. 

The variable is first calculated as the lagged year-over-year change, i.e., (realpcgdpUS[_n-1]-realpcgdpUS[_n-2])/realpcgdpUS[_n-2] in real-valued per capita GDP (in $US) in a given country. Then, winsorization is two-sided based on the fraction of 0.005 in each tail, based on the observations when main_sample==1. 


Variable: "chlgrealpcgdpUS"

lagged per capita GDP growth (in $US), unwinsorized. 

The variable is calculated as the lagged year-over-year change, i.e., (realpcgdpUS[_n-1]-realpcgdpUS[_n-2])/realpcgdpUS[_n-2] in real-valued per capita GDP (in $US) in a given country.  


Variable: "lnchlgrealpcgdpUS"

ln lagged per capita GDP growth (in $US); logged lagged per capita GDP growth (in $US)

The variable chlgrealpcgdpUS is transformed according to following formula: ln(chlgrealpcgdpUS+.53), based on the minimum value of "chlgrealpcgdpUS" when main_sample==1. 


Variable: "lnchlgrealpcgdpUSIHS"

lagged per capita GDP growth (in $US), IHS log-transformation. 

The variable chlgrealpcgdpUS is transformed according to following formula: ln((sqrt(chlgrealpcgdpUS^2)+1)+chlgrealpcgdpUS). 


Variable: "cpi"

Consumer price index (2010 = 100); series name "FP.CPI.TOTL" (Version Code: 201811) from World Development Indicators (see https://databank.worldbank.org/source/world-development-indicators)


Variable: "chcpi_w005e"
 
Inflation rate, winsorized main variable. 

The variable is first calculated as the year-over-year change (year[t]-year[t-1])/year[t-1] in consumer price index (2010=100) in a given country. Then, winsorization is two-sided based on the fraction of 0.005 in each tail, based on the observations when main_sample==1. 


Variable: "chcpi"

Inflation rate, unwinsorized. 

The variable is calculated as the year-over-year change (year[t]-year[t-1])/year[t-1] in consumer price index (2010=100) in a given country.   


Variable: "lnchcpi"

ln Inflation rate; logged inflation rate

The variable chcpi is transformed according to following formula: ln(chcpi+.08), based on the minimum value of "chcpi" when main_sample==1. 


Variable: "lnchcpiIHS"

Inflation rate, IHS log-transformation. 

The variable chcpi is transformed according to following formula: ln((sqrt(chcpi^2)+1)+chcpi). 


Variable: "realinterest"

Real interest rate (%) ; series code "FR.INR.RINR" (Version Code: 201811) from World Development Indicators (see https://databank.worldbank.org/source/world-development-indicators)


Variable: "chrealinterest_w005e"

Interest rate change, winsorized main variable.

The variable is first calculated as the change (year[t]-year[t-1]) in the real interest rate in a given country. Then, winsorization is two-sided based on the fraction of 0.005 in each tail, based on the observations when main_sample==1. 


Variable: "chrealinterest"

Interest rate change, unwinsorized.

The variable is first calculated as the change (year[t]-year[t-1]) in the real interest rate in a given country.  


Variable: "lgurbanpct" 

lagged % Urban ; lagged (1 year) series "SP.URB.TOTL.IN.ZS" (Version 202105) from World Development Indicators (see https://databank.worldbank.org/source/world-development-indicators)


Variable: "lgurbanpctelt"

Executive Election x lagged % Urban ; interaction between variables "elt" and "lgurbanpct" (elt * lgurbanpct)


Variable: "e_polity2"

Polity revised combined score ; variable "e_polity2" taken from V-Dem 10 dataset (Coppedge et al. 2020), originally from Polity IV (Marshall, Gurr, and Jaggers 2019).


Variable: "newfh"

Freedom in the World Status; Freedom House's (2020) score of political rights and civil liberties 

0: Not free
0.5: Partly free
1: Free


Variable: "president_fixed"

Presidential-Fixed Election System.

1: The country has a presidential or a semi-presidential system, where the president serves as the chief executive and there are fixed election schedules for the office of the presidency, and the president cannot call elections early.
0: otherwise


Variable: "main_sample"

Indicator to select the main sample of observations for all non-subset analyses. Takes into account data availability for the dependent and independent variables. Countries in sample are required to have at least three years of data and at least one election year and one non-election year. N.B. Observations in Burundi from 2003-2005 are excluded as the calculation of Private Fixed Investment Growth calculation is not possible due to values in t or t-1 being negative. 


Variable: "pfix1_sample"

Indicator to select the subset of observations for the "Presidential-Fixed Election System"==1 analysis. Takes into account data availability for the dependent and independent variables. Countries in sample are required to have at least three years of data and at least one election year and one non-election year. 


Variable: "pfix0_sample"

Indicator to select the subset of observations for the "Presidential-Fixed Election System"==0 analysis. Takes into account data availability for the dependent and independent variables. Countries in sample are required to have at least three years of data and at least one election year and one non-election year. 


##########################
Dataset: "POLCON_2017.dta"
##########################

The table includes the data from the file "POLCON_2017.xlsx" (Henisz 2017).


########### 
References: 
###########

Acemoglu, Daron, and Simon Johnson. 2005. "Unbundling Institutions." Journal of Political Economy 113(5): 949-995.

Acemoglu, Daron, Simon Johnson, and James A. Robinson. 2002. "Reversal of Fortune: Geography and Institutions in the Making of the Modern World Income Distribution." Quarterly Journal of Economics 117(4): 1231-1294.

Alesina, Alberto, Gerald D. Cohen, and Nouriel Roubini. 1993. "Electoral Business Cycles in Industrial Democracies." European Journal of Political Economy 9(1): 1-23.

Brender, Adi, and Allan Drazen. 2005. "Political Budget Cycles in New Versus Established Democracies." Journal of Monetary Economics 52(7): 1271-95.

Coppedge, Michael, John Gerring, Carl Henrik Knutsen, Staffan I. Lindberg, Jan Teorell, David Altman, Michael Bernhard, M. Steven Fish, Adam Glynn, Allen Hicken, Anna Lührmann, Kyle L. Marquardt, Kelly McMann, Pamela Paxton, Daniel Pemstein, Brigitte Seim, Rachel Sigman, Svend-Erik Skaaning, Jeffrey Staton, Steven Wilson, Agnes Cornell, Nazifa Alizada, Lisa Gastaldi, Haakon Gjerløw, Garry Hindle, Nina Ilchenko, Laura Maxwell, Valeriya Mechkova, Juraj Medzihorsky, Johannes von Römer, Aksel Sundström, Eitan Tzelgov, Yi-ting Wang, Tore Wig, and Daniel Ziblatt. 2020. "V-Dem Country–Year Dataset v10" Varieties of Democracy (V- Dem) Project. URL: https://doi.org/10.23696/vdemds20.

Cruz, Cesi, Philip Keefer, and Carlos Scartascini. 2018. "Database of Political Institutions 2017 (DPI2017)." Inter-American Development Bank. https://mydata.iadb.org/Reform- Modernization-of-the-State/Database-of-Political-Institutions-2017/938i-s2bw. 

Freedom House. 2020. Freedom in the World. URL: https://freedomhouse.org/report/freedom-
world (accessed 3/10/2020).

Gründler, K., and Krieger, T. (2021). "Using Machine Learning for measuring democracy: A practitioners guide and a new updated dataset for 186 countries from 1919 to 2019." European Journal of Political Economy, 70, 102047.

Henisz, Witold J. 2017. "The Political Constraint Index Dataset." Wharton School. URL: https://mgmt.wharton.upenn.edu/faculty/heniszpolcon/polcondataset/ (accessed 3/7/2020). 

Marshall, Monty G., Ted Robert Gurr, and Keith Jaggers. 2019. "Polity IV Project." Center for Systemic Peace.


